Comparison of Limulus amebocyte lysate test methods for endotoxin measurement in protein solutions.

The measurement of low levels of bacterial lipopolysaccharides (endotoxins) in protein samples can be a challenge due to potential interference from the inherent properties of the protein, its formulation or other substances that may be present. Other factors include the expression system which may have endotoxin species distinct from the standard, as well as different purification bioprocesses. The endotoxin measurement assays also have a number of variables. Those studied include differences between laboratories, reagents and standards, and detection modalities. A variety of protein samples from a range of expression systems was included in the evaluation. Endotoxin levels are relatively stable when samples are stored frozen with test variations between 1 and 38% among different aliquots. Test variation between labs was not significantly different when the same procedure was followed (intermediate precision) by trained analysts. Most testing modalities gave results within a 50-100% variation, a difference generally regarded as within assay variability. However, about 25% of the samples showed significant differences between testing modalities and/or reagents. The sources of these differences were further examined by traditional as well as novel sample treatments. These findings demonstrate that for some samples, endotoxin may be over- or under-estimated and a more thorough pre-treatment or testing modality may be required.